Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: sqlplus and a bug question

Re: sqlplus and a bug question

From: Michael Plawecki <m.plawecki_at_btinternet.com>
Date: Sat, 2 Sep 2000 21:06:49 +0100
Message-ID: <8ormnb$ccg$1@neptunium.btinternet.com>

I thought you had to do define _editor=vi in your sqlplus session (or in prefs.ora)?

Mike
Kirt Thomas <kremovethisspamthingthomas_at_gfsiinc.com> wrote in message news:c8k0rs47qtnapuqgoinn1qrm4lv168j3ni_at_4ax.com...
> First, in your environment set the EDITOR=/usr/bin/vi or whatever your
> path to vi is.... ie, at the prompt type;
>
> EDITOR=/usr/bin/vi;export EDITOR
>
> And voila !
>
> Second, have you tried metalink (metalink.oracle.com) - here's some of
> the bug text..
>
> This bug could result in server and background processes hanging in an
> idle-wait loop. Running truss on the process would show similar
> output as
> above, repeating the same sequence of system calls:
> semop (SIGALRM) sigprocmask times setitimer sigprocmask setcontext
> semop
> Note that the times() call would show a negative return value.
> This negative return value was interpreted as an error, causing the
> server
> process to continue the idle-wait loop and never time out.
> .
> If the machine clock ticks 100 times per second (truss -v would show
> HZ=100)
> then the times() call will return a negative value starting 248 days
> after the
> machine is booted. If the machine clock ticks 1000 times per second
> (HZ=1000),
> the times() call will return a negative value 24.8 days after the
> machine was
> booted.
> .
> A shutdown normal or immediate would hang as well. Only a shutdown
> abort could
> stop the instance, and it would hang again when mounting the
> database on
> startup.
> .
> The only recommended workaround for this problem is to reboot the
> machine
> where the times() call is returning a negative value, and schedule a
> reboot
> every 24 (HZ=1000) or 248 (HZ=100) days.
> .
> Although some customers were able to restart their database in this
> situation
> after recreating the controlfile, this is not the correct solution.
> You are
> strongly advised NOT to recreate your controlfile when this problem
> occurs. It
> may not help, and you will lose useful information stored in the
> controlfile.
>
>
>
>
> On Fri, 1 Sep 2000 19:04:27 -0400, "Legend" <legend_at_spacelab.net>
> wrote:
>
> >Hello all,
> >
> >How can I bring up vi editor within an sqlplus session? Also where can I
> >find out more information with an Oracle bug number #1084273?
> >
>
> o0o..o0o..o0o..o0o..o0o..o0o..o0o..o0o..o0o..o0o.o0o
> Kirt Thomas DBA- -kspamfreethomas_at_gfsiinc.com
Received on Sat Sep 02 2000 - 15:06:49 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US